1. The key knowledge points in this section are summed up in your own words, can be accompanied by pictures, and explain the importance of the knowledge points
Machine learning is mainly divided into these four categories:
The difference between regression and classification:
Definition of linear regression: a regression method that models between one or more independent variables and the dependent variable, which can be a linear combination between one or more independent variables.
Regression prediction is actually to find a weight multiplied by the eigenvalue plus the regression coefficient. y = w * x + b
There is a certain error between the machine prediction and the true value. We can reduce the error as much as possible, so there is a loss function (the size of the error).
Gradient descent of least squares method: With the increase of training times and continuous descent, the predicted value will be closer to the true value.
2. Thinking about what linear regression algorithms can be used for? (Everyone try not to write duplicates)
I think stocks can be predicted. Although there are many volatility factors in stocks, it is still possible to make certain predictions through data within a few years and certain volatility factors. There are funds, etc.
3. Write a linear regression algorithm independently, the data can be made by yourself, or obtained from the Internet. (Plus points)